Prognostic Factors for Progression of Neuroendocrine Tumors Identified in Reanalysis of RADIANT-2

James Yao, MD

The placebo-controlled phase III RADIANT-2 trial showed that the addition of everolimus to standard octreotide LAR achieved about a 5-month improvement in progression-free survival (PFS) in patients with advanced neuroendocrine tumors (NETs). At ASCO’s 2012 Gastrointestinal Cancers Symposium, a reanalysis of RADIANT-2 identified several significant prognostic factors associated with improved outcome: WHO performance status (PS), baseline chromogranin A (CgA), bone involvement, and lung as the primary tumor site.

“Prior to this study, because NET is such a rare disease, there hadn’t been good well-controlled studies to look at new treatments and understand the real prognostic factors. This population [of NET patients] is quite heterogeneous. When we looked at the placebo group we saw that some patients had PFS greater than 20 months while others had a much shorter PFS of around 8 months. The study helps us understand which patients with advanced NET are likely to progress in a short time,” said lead author James Yao, MD, from The MD Anderson Cancer Center, Houston, Texas. “A multivariate analysis adjusting for baseline imbalances in the two treatment arms found that everolimus is even more active than originally seen,” he added.

Octreotide LAR is the foundation of NET therapy. However, additional treatments are needed. Previous phase II studies suggested that everolimus, an mTOR inhibitor, had promising antitumor activity when combined with octreotide LAR in NET.

The phase III double-blind, placebo-controlled, trial randomized 423 patients with advanced NET to everolimus 10 mg/day plus octreotide LAR 30 mg every 28 days versus placebo plus the same dose and schedule of octreotide. Treatment was continued until disease progression, when patients in the placebo arm were allowed to cross over to the experimental arm. Median PFS was 16.4 months in the experimental arm versus 11.3 months with octreotide plus placebo, representing a 23% reduction in risk of events with the two-drug combination (P = .026).

“This difference in five-plus months favoring everolimus was ‘clinically meaningful,’” Yao stated.

The re-analysis addressed several imbalances in baseline characteristics of the trial favoring the placebo-plus-octreotide LAR arm: performance status, primary tumor in the lung, and previous use of chemotherapy.

A multivariate analysis identified the following factors as significant prognostic factors: treatment with everolimus (P = .003); WHO PS of 0 versus >1 (P = .006); elevated versus nonelevated CgA at baseline (P < .001); bone involvement versus no bone involvement (P = .02); and lung as primary site of cancer (P = .04).

“An exploratory analysis adjusted for these prognostic factors showed an even more significant benefit for everolimus therapy [P = .003],” Yao said. The relative risk reduction for disease progression improved from 23% in the original trial to 38% in the exploratory analysis. The benefit of the combination of everolimus and octreotide LAR was consistent across all subgroups.

Yao said that although CgA has been identified as a potential prognostic factor for PFS and OS, this is the first study to evaluate this biomarker in a randomized, placebo-controlled trial.

The most common side effects seen with combination treatment were stomatitis, rash, fatigue, and diarrhea, and these were mostly grades 1 and 2. Grade 3-4 adverse effects occurring with >5% frequency were stomatitis, fatigue, diarrhea, infections, and hyperglycemia.

Yao said that the study included patients with advanced NET, but everolimus is FDA-approved for only pancreatic NET, not other types of NET. He said that further studies are needed to expand the indication.